The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes standard developer tools, including git log and git grep, to search through repository history and source files for specific patterns associated with sensitive data. These commands are used for their intended purpose of security auditing.
[CREDENTIALS_UNSAFE]: The documentation includes various placeholder patterns and example strings (e.g., 'sk_live_xxxxx', 'AKIA[0-9A-Z]{16}') to illustrate how real secrets appear in code. These are used for educational and detection purposes and do not constitute a leak of actual credentials.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted data, specifically source code and JavaScript bundles, to find leaked secrets. Although this creates an ingestion point for external content, the skill's functionality is limited to pattern matching and reporting, which poses minimal risk of executing malicious instructions embedded in the analyzed data.
[EXTERNAL_DOWNLOADS]: The skill references established security resources and tools, such as the OWASP Secrets Management Cheat Sheet, GitHub Secret Scanning documentation, and the git-secrets utility. These references are to well-known, trusted industry sources used for remediation guidance.

leaked-secrets